| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 80.0, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.9692208514878444e-05, | |
| "loss": 2.4048, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.877641290737884e-05, | |
| "loss": 1.6571, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 4.72751631047092e-05, | |
| "loss": 1.088, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 4.522542485937369e-05, | |
| "loss": 0.7735, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 4.267766952966369e-05, | |
| "loss": 0.6017, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 3.969463130731183e-05, | |
| "loss": 0.4956, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "learning_rate": 3.634976249348867e-05, | |
| "loss": 0.4105, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "learning_rate": 3.272542485937369e-05, | |
| "loss": 0.3463, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "learning_rate": 2.8910861626005776e-05, | |
| "loss": 0.2862, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2345, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "learning_rate": 2.1089138373994223e-05, | |
| "loss": 0.1905, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "learning_rate": 1.7274575140626318e-05, | |
| "loss": 0.1554, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "learning_rate": 1.3650237506511331e-05, | |
| "loss": 0.1269, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "learning_rate": 1.0305368692688174e-05, | |
| "loss": 0.1071, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "learning_rate": 7.3223304703363135e-06, | |
| "loss": 0.0921, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "learning_rate": 4.7745751406263165e-06, | |
| "loss": 0.0832, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "learning_rate": 2.7248368952908053e-06, | |
| "loss": 0.0777, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "learning_rate": 1.2235870926211619e-06, | |
| "loss": 0.0739, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "learning_rate": 3.077914851215585e-07, | |
| "loss": 0.0714, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.072, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "step": 200, | |
| "total_flos": 3.758484756417741e+16, | |
| "train_loss": 0.4674161618947983, | |
| "train_runtime": 1650.9942, | |
| "train_samples_per_second": 4.846, | |
| "train_steps_per_second": 0.121 | |
| } | |
| ], | |
| "max_steps": 200, | |
| "num_train_epochs": 100, | |
| "total_flos": 3.758484756417741e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |